Antibody (Ig) molecules produced by B-lymphocytes are built of heavy (H) and light (L) chains. The amino acid sequences of the amino terminal domains of the H and L chains are variable (VH and VL), especially at the three hypervariable regions (CDR1, CDR2, CDR3) that form the antigen combining site. The assembly of the H and L chains is stabilized by a disulfide bond between the constant region of the L chain (CL) and the first constant region of the heavy chain (C.sub.H1) and by non-covalent interactions between the VH and VL domains.
In humans and many animals, such as mice, the genes encoding the antibody H and L chains are assembled by stepwise somatic rearrangements of gene fragments encoding parts of the V regions. Various stages of B lymphocyte development are characterized by the rearrangement status of the Ig gene loci (see, e.g. Melchers, F. & Rolink, A., B-Lymphocyte Development and Biology, Paul, W. E., ed., 1999, Lippincott, Philadephia).
Precursors of B cells (pre-B cells) have been identified in the bone marrow as lymphocytes that produce μ heavy chains but instead of the fully developed light chains express a set of B lineage-specific genes called VpreB(1-3) and λ5, respectively.
The main isoform of human VpreB1 (CAG30495) is a 145 aa-long polypeptide (SEQ ID NO: 1). It has an Ig V domain-like structure, but lacks the last β-strand (β7) of a typical V domain, and has a carboxyl terminal end that shows no sequence homologies to any other proteins. VpreB2 has several isoforms, including a 142-amino acid mouse VpreB2 polypeptide (P13373; SEQ ID NO: 2), and a 171-amino acid long splice variant of the mouse VPreB2 sequence (CAA019641 SEQ ID NO: 3). VpreB1 and VpreB2 sequences have been disclosed in EP 0 269 127 and U.S. Pat. No. 5,182,205; Collins et al., Genome Biol. 5(10):R84 (2004); and Hollins et al., Proc. Natl. Acad. Sci. USA 86(14):5552-5556 (1989). The main isoform of human VpreB3 (SEQ ID NO: 4) is a 123 amino acid long protein (CAG30496), disclosed in Collins et. al., Genome Biol. 5(10):R84 (2004).
VpreB(1-3) are non-covalently associated with another protein, λ5. The human λ5 is a 209-amino acid polypeptide (CAA01962; SEQ ID NO: 5), that carries an Ig C domain-like structure with strong homologies to antibody light chains and, towards its amino terminal end, two functionally distinct regions, one of which shows strong homology to the β7 strand of the Vλ domains. A human λ5-like protein has 213 amino acids (NP_064455; SEQ ID NO: 6) and shows about 84% sequence identity to the antibody λ light chain constant region.
For further details, see the following review papers: Karasuyama et al., Adv. Immunol. 63:1-41 (1996); Melchers et al., Immunology Today 14:60-68 (1993); and Melchers, Proc. Natl. Acad. Sci. USA 96:2571-2573 (1999).
The VpreB and λ5 polypeptides together form a non-covalently associated, Ig light chain-like structure, which is called the surrogate light chain or pseudo light chain. On the surface of early preB cells, the surrogate light chain is disulfide-linked to membrane-bound Ig μ heavy chain in association with a signal transducer CD79a/CD79b heterodimer to form a B cell receptor-like structure, the so-called preB cell receptor (preBCR).